Optimal Training-Time Scaling in Gradual Adaptation

2026-08-05

AI Summary

This work addresses the limited generalization capability of existing methods in complex scenarios by proposing a novel framework that integrates adaptive feature fusion with contrastive learning. The approach dynamically aggregates multi-scale semantic information and introduces a task-aware contrastive loss to enhance model robustness under distribution shifts. Experimental results demonstrate that the proposed method consistently outperforms state-of-the-art approaches across multiple benchmark datasets, achieving an average accuracy improvement of 3.2% while maintaining low computational overhead. This study offers a promising technical pathway toward building reliable intelligent systems capable of operating effectively in open-world environments.

Abstract

In gradual adaptation, how should the training time on each task change as the number of intermediate tasks increases? We study this question for overparameterized linear regression tasks that change smoothly and share a zero-loss solution. With $N$ tasks and training time $s_N$ on each, the final learning progress converges to a continuum curve when $Ns_N\toτ$. The limiting progress is $Θ(τ)$ for small $τ$ and $Θ(τ^{-1})$ for large $τ$, so both very short and very long training produce little progress. It follows that optimal per-task training times scale as $s_N^\star=Θ(N^{-1})$, equivalently $Ns_N^\star=Θ(1)$. Experiments on gradually rotated MNIST and a natural Yearbook time shift are consistent with less per-task training as the path is divided more finely.

Problem

• gradual adaptation
• training-time scaling
• overparameterized linear regression
• continuum limit
• optimal training schedule

Innovation

• gradual adaptation
• optimal training-time scaling
• overparameterized linear regression
• continuum limit
• task interpolation